Product developers require test data with which to test products and services under development. It is important that this data be of such content and form that when used to test a new product, any errors generated during testing are attributed to failure of the product, not to erroneous test data. In the absence of error-free test data, a failure during testing of a new product requires an investigation to determine whether the product simply needs additional work or that the test data was bad. Such an investigation may take significant time, and therefore, increase development costs.
Many enterprises define an internal data management group (“Data Management”), consisting of an individual or individuals, which is responsible for providing test data to product developers. Data Management typically receives requests for test data from product developers and responds to each request on an individual first come, first serve basis. Data Management may fill each request by copying data from a source or sources and transmitting a text file containing that data to the requesting product developer by email. Distributing test data in this manner is inefficient for a number of reasons. First, Data Management copies test data from its source each time a request is made. Second, with each request, test data is converted to a text file and transmitted to the requesting product developer. Third, because of the overhead involved with requesting test data, product developers often request more test data than needed so as to avoid the need to submit a second request, should they require more test data. These larger-than-necessary requests for test data burden the already inefficient distribution method.
In addition to being inefficient, this method of distributing test data has other shortcomings. Distributing test data in text files transmitted via email is an unsecure method of sharing data. Unintended recipients, and/or hackers, may gain access to this data, which may include sensitive, even confidential, customer information. Also, test data distributed in this manner has not been validated, meaning it may contain erroneous data which when used may cause failures during testing and, as described above, increase development costs.